Roles of microphysics in cloud resolving models in passive
microwave remote sensing of precipitation over ocean
1Ju-Hye
Kim,
1Korea
2Dong-Bin
Shin, and
3Christian
Kummerow
Institute of Atmospheric Prediction Systems, Seoul, Korea / j.kim@kiaps.org
2Yonsei University, Seoul, Korea
3Colorado State University, CO, U.S.A.
1. Remote sensing of precipitation
2. Uncertainty of microwave remote
sensing of precipitation
 Microwave sensors are known as
the most accurate precipitation
measurement owing to its
physical relationship with
precipitation particles.
 Forward model error : Radiative transfer
model, surface variability, bulk microphysics
 Frozen hydrometeors scatter
upwelling radiations while cloud
and rain droplets emit
microwave radiation. These two
properties are distinct depending
on the frequency.
Observed brightness temperatures (TBs) for typhoon Sudal (2004) from TRMM
satellite measurement.
Most of current microwave rainfall
algorithm depends on the cloud
resolving model (CRM) in the
generation of a-priori DB!
 The largest uncertainty is introduced by the
cloud model itself. Assumptions in the
microphysical properties and related
characteristics of hydrometeors distributions
are connected to microwave rainfall retrieval
errors.
3. Parametric rainfall algorithm (GPROF v7 Ocean algorithm)
1) Generate DBs
λ5
Z= 5 2
π K
∫σ
D
b
To overcome the dependency on
the CRM, the ‘Parametric rainfall
algorithm’ was developed.
2) Inversion by Bayesian theory
The key of this methodology is
generating a-priori rainfall DB
using TRMM PR observation, it is
the PR constrained DB.
( D) N ( D)dD
PIA = 20τ ext (0, ∞) log e
Selected model
hydrometeors profile
P (R | Tb ) ∝ P (R ) × P (Tb | R )
Observed/Simulated
reflectivity profile
Expected RR by this
observed 9 TBs set for
this rain rate




Probability of Inversely weighted by
this rain rate differences of observed
and simulated 9 TBs for
occurrence
each profiles
Shin and Kummerow (2003)
Masunaga and Kummerow (2005)
Elsaesser and Kummerow (2008)
Kummerow et al. (2011)
4. Part I: Impact of a-priori DBs using six WRF microphysics schemes
Vertical distribution of time- and
area- averaged hydrometeors from
six simulations
 More cloud and rain water
and more ice particle than
the WSM6
Distribution of observed TBs for databases
and retrieval target
Experimental Design
Rainfall retrieval results
 Similar distribution of
rain and cloud water
compare to the
WSM6
 Reduction of snow
near and above the
melting layer
2A12 (GPROF v7)
2A25 (PR)
Corr= 0.62
RMS= 12.07
Bias= -2.28
 Too much snow
 Relatively less rain
water
TBs for build DBs (36522+36532)
TBs for retrievals (36537)
 Increased rain water
below 5 km altitude
 Similar distribution of ice
particle compare to the
WSM6
Six cloud microphysics in WRF model
Single M
Bulk
PLIN
Lin et al. 1983
Chen and Sun 20021
WSM6
Hong et al. 2004
Hong and Lim 2006
qv, qc, qi, qr, qs, qg
Tao and Simpson 1993
GCE
Lang et al. 2007
Reisner et al. 1998
THOM
Double M
Bulk
WDM6
Thompson et al. 2004, 2008
Lim and Hong 2010
Reisner et al. 1998
MORR
qv, qc, qi,
qr,
Ncw, Nrw, NCCN
qs, qg
Nci, Nsi
Ncw, Nci, Ncr, Ncs
Morrison et al. 2005
Single moment
schemes have
differences in their cold
rain processes (ice
initiation,
sedimentation property
of solid particles).
Cold rain
Double M for
Warm rain
Warm rain , Cold rain
As expected, the characteristics of the a-priori databases are inherited from the individual cloud
microphysics schemes. Major results show that convective rainfall regions are not well captured
by the LIN and THOM schemes-based retrievals. Rainfall distributions and their quantities
retrieved from the WSM6 and WDM6 schemes-based estimations, however, show relatively
better agreement with the PR observations.
Corr= 0.56
RMS= 11.98
Bias= -1.32
Corr= 0.79
RMS= 8.87
Bias= -0.56
Corr= 0.78
RMS= 9.14
Bias= -0.61
Corr= 0.73
RMS= 9.77
Bias= -0.20
Corr= 0.81
RMS= 8.64
Bias= -0.42
Corr= 0.78
RMS= 9.43
Bias= -0.81
Kim, J.-H., D.-B., Shin, and C. D. Kummerow, 2013: Impacts of a-priori databases using six WRF microphysics schemes on passive microwave rainfall retrievals. J. Atmos. Oceanic Technol., 30, 2367–2381.
5. Part II: Two heavy rainfall cases with different cloud microphysics
This study includes the discrepancy of estimated rain rate from passive radiometer and active radar for two rainfall systems of
different cloud microphysics near the Yellow Sea. The first case have high cloud top (HCT, cold type) with large ice particles and
the other case is precipitation with mid cloud top (MCT, warm type) having less ice particles.
Sensitivity experiment by
various a priori DBs
Case 2: Warm type precipitation
Observed
In 10 GHz
channel, emission
from rain water is
significant.
Simulated component
Observed component
Simulated component
TRMM observations for two rainfall cases
Case 1: Cold type precipitation
ex) 4W_2004MJJAS
In 19 GHz channel,
emission from rain
and cloud water is
significant.
component
4 WRF CRMs
5 WRF CRMs
9 GPROF CRMs
2004 MJJAS
4W_2004MJJAS
5W_2004MJJAS
9G_2004MJJAS
2004 JJA
4W_2004JJA
5W_2004JJA
9G_2004JJA
2005 JJA
4W_2005JJA
5W_2005JJA
9G_2005JJA
2004 JJA, 2005 JJA
4W_0405JJA
5W_0405JJA
9G_0405JJA
4W_0405mJJAs
5W_0405mJJAs
9G_0405mJJAs
2004 MJJAS,
2005 JJA
The 85 GHz channel
is very sensitive
scattering from ice
particles located in
upper clouds.
The 37 GHz
channel has both
emission (cloud
particles) and
scattering signals
(large ice
particles).
Large
scattering!
Max. rain rate :
above 60 mm/h
Max. rain rate :
above 60 mm/h
Max. rain rate :
above 10 mm/h
Less
scattering!
(little ice
particle)
Max. rain rate :
above 40 mm/h
Findings and concluding remarks
 The dependency of CRM and its microphysics is significant in rainfall estimation from deep convective cloud, because this system has ice particles in the upper
atmosphere which brings a scattering signal at high frequency channels. On the other hand, high resolution TBs of 85 GHz (7 km x 5 km) channel gives little
knowledge in warm type precipitation case and it brings underestimation of rain rate.
 In GPM era, a network of microwave radiometers to provide the best possible global coverage and sampling including high latitude regions, where complicate
microphysics is revealed. I hope this research could contribute to improve accuracy of microwave rainfall measurements for various cloud and precipitation
systems by understanding the role of microphysics in CRMs in microwave rainfall estimation.
19th International TOVS working group Study Conferences | 26 March- 1 April 2014, JeJu, Korea
Image Credit: NASA/Bill Ingalls
To better understand Earth's weather and climate
cycles, the GPM Core Observatory will collect
information that unifies and improves data from
an international constellation of existing and
future satellites by mapping global precipitation
every three hours.
Fri, 2014-03-20 9:41 a.m. EDT
The Global Precipitation Measurement mission's
Core Observatory is performing normally. Both
the GPM Microwave Imager and Dual-frequency
Precipitation Radar are collecting science data.
Thu, 2014-02-27 1:38 p.m. EST
Liftoff! The GPM Core Observatory launched on
Feb. 27 2014 (above).